A survey of cultures by the anthropologist Laila Williamson reveals that infanticide has been practiced on every continent and by every kind of society, from nonstate bands and villages (77 percent of which have an accepted custom of infanticide) to advanced civilizations. Until recently, between 10 and 15 percent of all babies were killed shortly after they were born, and in some societies the rate has been as high as 50 percent. In the words of the historian Lloyd deMause, "All families once practiced infanticide. All states trace their origin to child sacrifice. All religions began with the mutilation and murder of children."

What do Moses, Ishmael, Romulus and Remus, Oedipus, Cyrus the Great, Sargon, Gilgamesh, and Hou Chi (a founder of the Chou Dynasty) have in common? They were all exposed as infants -- abandoned by their parents and left to the elements. The image of a helpless baby dying alone of cold, hunger, and predation is a potent tug on the heartstrings, so it is not surprising that a rise from infant exposure to dynastic greatness found its way into the mythologies of Jewish, Muslim, Roman, Greek, Persian, Akkadian, Sumerian, and Chinese civilizations. But the ubiquity of the exposure archetype is not just a lesson in what makes for a good story arc. It is also a lesson on how common infanticide was in human history. From time immemorial, parents have abandoned, smothered, strangled, beaten, drowned, or poisoned many of their newborns.

How can we make sense of something that runs as contrary to the continuation of life as killing a newborn? In the concluding chapter of Hardness of Heart/Hardness of Life, his magisterial survey of infanticide around the world, the physician Larry Milner makes a confession:

Though infanticide is the most extreme form of maltreatment of children, our cultural heritage tells of many others, including the sacrifice of children to gods; the sale of children into slavery, marriage, and religious servitude; the exploitation of children to clean chimneys and crawl through tunnels in coal mines; and the subjection of children to forms of corporal punishment that verge on or cross over into torture. We have come a long way to arrive at an age in which one-pound preemies are rescued with heroic surgery, children are not expected to be economically productive until their fourth decade, and violence against children has been defined down to dodgeball.

I began this book with one purpose in mind -- to understand, as stated in the Introduction: "How someone can take their own child, and strangle it to death?" When I first raised the question many years ago, I thought the issue to be suggestive of some unique pathologic alteration of Nature's way. It did not seem rational that evolution would maintain an inherited tendency to kill one's offspring when survival was already in such a delicate balance. Darwinian natural selection of genetic material meant that only the survival of the fittest was guaranteed; a tendency toward infanticide must certainly be a sign of unfit behavior that would not pass this reasonable standard. But the answer which has emerged from my research indicates that one of the most "natural" things a human being can do is voluntarily kill its own offspring when faced with a variety of stressful situations.

The solution to Milner's puzzlement lies in the subfield of evolutionary biology called life history theory. The intuition that a mother should treat every offspring as infinitely precious, far from being an implication of the theory of natural selection, is incompatible with it. Selection acts to maximize an organism's expected lifetime reproductive output, and that requires that it negotiate the tradeoff between investing in a new offspring and conserving its resources for current and future offspring. Mammals are extreme among animals in the amount of time, energy, and food they invest in their young, and humans are extreme among mammals. Pregnancy and birth are only the first chapter in a mother's investment career, and a mammalian mother faces an expenditure of more calories in suckling the offspring to maturity than she expended in bearing it. Nature generally abhors the sunk-cost fallacy, and so we expect mothers to assess the offspring and the circumstances to decide whether to commit themselves to the additional investment or to conserve their energy for its born or unborn siblings. If a newborn is sickly, or if the situation is unpromising for its survival, they do not throw good money after bad but cut their losses and favor the healthiest in the litter or wait until times get better and they can try again.

To a biologist, human infanticide is an example of this triage. Until recently, women nursed their children for two to four years before returning to full fertility. Many children died, especially in the perilous first year. Most women saw no more than two or three of their children survive to adulthood, and many did not see any survive. To become a grandmother in the unforgiving environment of our evolutionary ancestors, a woman would have had to make hard choices. The triage theory predicts that a mother would let a newborn die when its prospects for survival to adulthood were poor. The forecast may be based on bad signs in the infant, such as being deformed or unresponsive, or bad signs for successful motherhood, such as being burdened with older children, beset by war or famine, or unable to count on support from relatives or the baby's father. It should also depend on whether she is young enough to have opportunities to try again.

Martin Daly and Margo Wilson tested the triage theory by examining a sample of sixty unrelated societies from a database of ethnographies. Infanticide was documented in a majority of them, and in 112 cases the anthropologists recorded a reason. Eighty-seven percent of the reasons fit the triage theory: the infant was not sired by the woman's husband, the infant was deformed or ill, or the infant had strikes against its chances of surviving to maturity, such as being a twin, having an older sibling close in age, having no father around, or being born into a family that had fallen on hard economic times.

The ubiquity and evolutionary intelligibility of infanticide suggest that for all its apparent inhumanity, it is usually not a form of wanton murder but falls into a special category of violence. Anthropologists who interview these women (or their relatives, since the event may be too painful for the woman to discuss) often recount that the mother saw the death as an unavoidable tragedy and grieved for the lost child. Napoleon Chagnon, for example, wrote of the wife of a Yanomamö headman, "Bahami was pregnant when I began my fieldwork, but she destroyed the infant when it was born -- a boy in this case -- explaining tearfully that she had no choice. The new baby would have competed with Ariwari, her youngest child, who was still nursing. Rather than expose Ariwari to the dangers and uncertainty of an early weaning, she chose to terminate the newborn instead." Though the Yanomamö are the so-called fierce people, infanticide is not necessarily a manifestation of fierceness across the board. Some warring tribes, particularly in Africa, rarely kill their newborns, while some relatively peaceful ones kill them regularly. The title of Milner's magnum opus comes from a quotation from a 19th-century founder of anthropology, Edward Tylor, who wrote, "Infanticide arises from hardness of life rather than hardness of heart."

The fateful tipping point between keeping and sacrificing a newborn is set both by internal emotions and by cultural norms. In a culture such as ours that reveres birth and takes every step to allow babies to thrive, we tend to think that joyful bonding between mother and newborn is close to reflexive. But in fact it requires overcoming considerable psychological obstacles. In the 1st century CE, Plutarch pointed out an uncomfortable truth:

There is nothing so imperfect, so helpless, so naked, so shapeless, so foul, as man observed at birth, to whom alone, one might almost say, Nature has given not even a clean passage to the light; but, defiled with blood and covered with filth, and resembling more one just slain than one just born, he is an object for none to touch or lift up or kiss or embrace except for someone who loves with a natural affection.

The "natural affection" is far from automatic. Daly and Wilson, and later the anthropologist Edward Hagen, have proposed that postpartum depression and its milder version, the baby blues, are not a hormonal malfunction but the emotional implementation of the decision period for keeping a child. Mothers with postpartum depression often feel emotionally detached from their newborns and may harbor intrusive thoughts of harming them. Mild depression, psychologists have found, often gives people a more accurate appraisal of their life prospects than the rose-tinted view we normally enjoy. The typical rumination of a depressed new mother -- how will I cope with this burden?-- has been a legitimate question for mothers throughout history who faced the weighty choice between a definite tragedy now and the possibility of an even greater tragedy later. As the situation becomes manageable and the blues dissipate, many women report falling in love with their baby, coming to see it as a uniquely wonderful individual.

Many cultural traditions work to distance people's emotions from a newborn until its survival seems likely. People may be enjoined from touching, naming, or granting legal personhood to a baby until a danger period is over, and the transition is often marked by a joyful ceremony, as in our own customs of the christening and the bris. Some traditions have a series of milestones, such as traditional Judaism, which grants full legal personhood to a baby only after it has survived thirty days.

Hagen examined the psychiatric literature on postpartum depression to test five predictions of the theory that it is an evaluation period for investing in a newborn. As predicted, postpartum depression is more common in women who lack social support (they are single, separated, dissatisfied with their marriage, or distant from their parents), who had had a complicated delivery or an unhealthy infant, and who were unemployed or whose husbands were unemployed. He found reports of postpartum depression in a number of non-Western populations which showed the same risk factors (though he could not find enough suitable studies of traditional kin-based societies). Finally, postpartum depression is only loosely tied to measured hormonal imbalances, suggesting that it is not a malfunction but a design feature.

If I have tried to make infanticide a bit more comprehensible, it is only to reduce the distance between the vast history in which it was accepted and our contemporary sensibilities in which it is abhorrent. But the chasm that separates them is wide. Even when we acknowledge the harsh evolutionary logic that applies to the hard lives of premodern peoples, many of their infanticides are, by our standards, hard to comprehend and impossible to forgive. Examples from Daly and Wilson's list include the killing of a newborn conceived in adultery, and the killing of all a woman's children from a previous marriage when she takes (or is abducted by) a new husband. And then there are the 14 percent of the infanticidal justifications on the list that, as Daly and Wilson point out, do not easily fall into categories that an evolutionary biologist would have predicted beforehand. They include child sacrifice, an act of spite by a grandfather against his son-in-law, filicides that are committed to eliminate claimants to a throne or to avoid the obligations of kinship customs, and most commonly, the killing of a newborn for no other reason than that she is a girl.

Female infanticide has been put on the world's agenda today by census data revealing a massive shortage of women in the developing world. "A hundred million missing" is the commonly cited statistic for the daughter shortfall, a majority of them in China and India. Many Asian families have a morbid preference for sons. In some countries a pregnant woman can walk into an amniocentesis or ultrasound clinic, and if she learns she is carrying a girl, she can walk next door to an abortion clinic. The technological efficiency of daughter-proofing a pregnancy may make it seem as if the girl shortage is a problem of modernity, but female infanticide has been documented in China and India for more than two thousand years. In China, midwives kept a bucket of water at the bedside to drown the baby if it was a girl. In India there were many methods: "giving a pill of tobacco and bhang to swallow, drowning in milk, smearing the mother's breast with opium or the juice of the poisonous Datura, or covering the child's mouth with a plaster of cow-dung before it drew breath." Then and now, even when daughters are suffered to live, they may not last long. Parents allocate most of the available food to their sons, and as a Chinese doctor explains, "if a boy gets sick, the parents may send him to the hospital at once, but if a girl gets sick, the parents may say to themselves, 'Well, we'll see how she is tomorrow.'"

Female infanticide is biologically mysterious. Every child has a mother and a father, so if people are concerned about posterity, be it for their genes or their dynasty, culling their own daughters is a form of madness. A basic principle of evolutionary biology is that a fifty-fifty sex ratio at sexual maturity is a stable equilibrium in a population, because if males ever predominated, daughters would be in demand and would have an advantage over sons in attracting partners and contributing children to the next generation. And so it would be for sons if females ever predominated. To the extent that parents can control the sex ratio of their surviving offspring, whether by nature or by nurture, posterity should punish them for favoring sons or daughters across the board.

Female infanticide, also called gendercide and gynecide, is not unique to Asia. The Yanomamö are one of many foraging peoples that kill more newborn daughters than sons. In ancient Greece and Rome, babies were "discarded in rivers, dunghills, or cesspools, placed in jars to starve, or exposed to the elements and beasts in the wild." Infanticide was also common in medieval and Renaissance Europe. In all these places, more girls perished than boys. Often families would kill every daughter born to them until they had a son; subsequent daughters were allowed to live.

One naïve hypothesis comes out of the realization that it is the number of females in a population that determines how rapidly it will grow. Perhaps tribes or nations that have multiplied themselves to the Malthusian limit on food or land kill their daughters to achieve zero population growth. One problem for the ZPG theory, however, is that many infanticidal tribes and civilizations were not environmentally stressed. A more serious problem is that it has the fatal flaw of all naïve good-of-the-group theories, namely that the mechanism it proposes is self-undermining. Any family that cheated on the policy and kept its daughters alive would take over the population, stocking it with their grandchildren while the excess bachelor sons of their altruistic neighbors died without issue. The lineages that were inclined to kill their newborn daughters would have died out long ago, and the persistence of female infanticide in any society would be a mystery.

Can evolutionary psychology explain the gender bias? Critics of that approach say that it is merely an exercise in creativity, since one can always come up with an ingenious evolutionary explanation for any phenomenon. But that is an illusion, arising from the fact that so many ingenious evolutionary hypotheses have turned out to be confirmed by the data. Such success is far from guaranteed. One prominent hypothesis that, for all its ingenuity, turned out to be false was the application of the Trivers-Willard theory of sex ratios to female infanticide in humans.

The biologist Robert Trivers and the mathematician Dan Willard reasoned that even though sons and daughters are expected to yield the same number of grandchildren on average, the maximum number that each sex can promise is different. A superfit son can outcompete other males and impregnate any number of women and thereby have any number of children, whereas a superfit daughter can have no more than the maximum she can bear and nurture in her reproductive career. On the other hand a daughter is a safer bet -- an unfit son will lose the competition with other men and end up childless, whereas an unfit daughter almost never lacks for a willing sex partner. It's not that her fitness is irrelevant -- a healthy and desirable daughter will still have more surviving children than an unhealthy and undesirable one -- but the difference is not as extreme as it is for boom-or-bust sons. To the extent that parents can predict the fitness of their children (say, by monitoring their own health, nutrition, or territory) and strategically tilt the sex ratio, they should favor sons when they are in better shape than the competition, and favor daughters when they are in worse shape.

The Trivers-Willard theory has been confirmed in many nonhuman species and even, a bit, in Homo sapiens. In traditional societies, richer and higher-status people tend to live longer and attract more and better mates, so the theory predicts that higher-status people should favor sons and lower-status people should favor daughters. In some kinds of favoritism (like bequests in wills), that is exactly what happens. But with a very important kind of favoritism -- allowing a newborn to live -- the theory doesn't work so well. The evolutionary anthropologists Sarah Hrdy and Kristen Hawkes have each shown that the Trivers-Willard theory gets only half of the story right. In India, it's true that the higher castes tend to kill their daughters. Unfortunately, it's not true that the lower castes tend to kill their sons. In fact, it's hard to find a society anywhere that kills its sons. The infanticidal cultures of the world are either equal-opportunity baby-killers or they prefer to kill the girls -- and with them, the Trivers-Willard explanation for female infanticide in humans.

The ultimate misogyny of female infanticide may suggest a feminist analysis, in which the sexism of a society extends to the right to life itself: being female is a capital offense. But that hypothesis doesn't work either. No matter how sexist these societies were (or are), they did not want a world that was Frauenfrei. The men do not live in all-boy treehouses in which no girls are allowed; they depend on women for sex, children, child-rearing, and the gathering or preparation of most of their food. Families that kill their daughters want there to be women around. They just want someone else to raise them. Female infanticide is a kind of social parasitism, a free rider problem, a genealogical tragedy of the commons.

Free rider problems arise when no one owns a common resource, in this case, the pool of potential brides. In a free market in marriages in which parents wielded property rights, sons and daughters would be fungible, and neither sex would be favored across the board. If you really needed a fierce warrior or brawny field hand around the house, it shouldn't matter whether you raised a son for the job or raised a daughter who would bring you a son-in-law. Families with more sons would trade some of them for daughters-in-law and vice versa. True, your son-in-law's parents might prefer him to stay with them, but you could use your bargaining position to force the young man to move in with you if he wanted to have a wife at all. A preference for sons should arise only in a market with distorted property rights, one in which parents, in effect, own their sons but not their daughters.

What about state societies like India and China? In state societies that practice female infanticide, Hawkes noted, parents also own their sons but not their daughters, though for economic rather than military reasons. In stratified societies whose elites have indivisible wealth, the inheritance often goes to a son. In India the caste system was an additional market distorter: lower castes had to pay steep dowries so their daughters could marry higher-caste grooms. In China, parents had a permanent claim on the support of their sons and daughters-in-law extending into their dotage, but not of their daughters and sons-in-law (hence the traditional adage "A daughter is like spilled water.") China's one-child policy, introduced in 1978, made parents' need for a son to support them in their old age all the more acute. In all these cases, sons are an economic asset and daughters a liability, and parents respond to the distorted incentives with the most extreme measures. Today infanticide is illegal in both countries. In China, infanticide is thought to have given way to sex-selective abortions, which are also illegal, though still widely practiced. In India, despite the inroads of the ultrasound-abortion chains, it is thought to remain common. The pressure to reduce these practices will almost certainly increase, if only because governments have finally done the demographic arithmetic and realized that gynecide today means unruly bachelors tomorrow (a phenomenon we will revisit).

Hawkes noted that among foraging peoples, female infanticide is more common in patrilocal societies (in which daughters move away to live with their husbands and in-laws) than in matrilocal societies (in which they stay with their parents and their husbands move in with them) or in societies in which the couple goes wherever they want. Patrilocal societies are common in tribes in which neighboring villages are constantly at war, which encourages related men to stay and fight together. They are less common when the enemies are other tribes, and the men have more freedom of movement within their territory. The internally warring societies then fall into a vicious circle in which they kill their newborn girls so their wives can hurry up and bear them more warrior sons, the better to raid other villages, and to defend their own villages against being raided, for a supply of women that had been decimated by their own infanticides. The warring tribes in Homeric Greece were caught in a similar trap.

Whether they are new mothers in desperate straits, putative fathers doubting their paternity, or parents preferring a son over a daughter, people in the West can no longer kill their newborns with impunity. In 2007 in the United States, 221 infants were murdered out of 4.3 million births. That works out to a rate of 0.00005, or a reduction from the historical average by a factor of two to three thousand. About a quarter of them were killed on their first day of life by their mothers, like the "trash-can moms' who made headlines in the late 1990s by concealing their pregnancies, giving birth in secret (in one case during a high school prom), smothering their newborns, and discarding their bodies in the trash. These women find themselves in similar conditions to those who set the stage for infanticide in human prehistory: they are young, single, give birth alone, and feel they cannot count on the support of their kin. Other infants were killed by fatal abuse, often by a stepfather. Still others perished at the hands of a depressed mother who committed suicide and took her children with her because she could not imagine them living without her. Rarely, a mother with postpartum depression will cross the line into postpartum psychosis and kill her children under the spell of a delusion, like the infamous Andrea Yates, who in 2001 drowned her five children in a bathtub.

What drove down the Western rate of infanticide by more than three orders of magnitude? The first step was to criminalize it. Biblical Judaism prohibited filicide, though it didn't go the whole hog: killing an infant younger than a month did not count as murder, and loopholes were claimed by Abraham, King Solomon, and Yahweh himself for Plague #10. The prohibition became clearer in Talmudic Judaism and in Christianity, from which it was absorbed into the late Roman Empire. The prohibition came from an ideology that held that lives are owned by God, to be given and taken at his pleasure, so the lives of children no longer belonged to their parents. The upshot was a taboo in Western moral codes and legal systems on taking an identifiable human life: one could not deliberate on the value of the life of an individual in one's midst. (Exceptions were exuberantly made, of course, for heretics, infidels, uncivilized tribes, enemy peoples, and transgressors of any of several hundred laws. And we continue to deliberate on the value of statistical lives, as opposed to identifiable lives, every time we send soldiers or police into harm's way, or scrimp on expensive health and safety measures.)

The victim's mind is not sufficiently developed to enable it to suffer from the contemplation of approaching suffering or death. It is incapable of feeling fear or terror. Nor is its consciousness sufficiently developed to enable it to suffer pain in appreciable degree. Its loss leaves no gap in any family circle, deprives no children of their breadwinner or their mother, no human being of a friend, helper, or companion.

Today we know that infants do feel pain, but in other ways Mercier's line of reasoning has been taken up by several contemporary philosophers -- invariably pilloried when their essays are brought to light -- who have probed the shadowy regions of our ethical intuitions in cases of abortion, animal rights, stem cell research, and euthanasia. And while few people would admit to observations like Mercier's, they creep into intuitions that in practice distinguish the killing of a newborn by its mother from other kinds of homicide. Many European legal systems separate the two, defining a separate crime of infanticide or neonaticide, or granting the mother a presumption of temporary insanity. Even in the United States, which makes no such distinction, when a mother kills a newborn prosecutors often don't prosecute, juries rarely convict, and those found guilty often avoid jail. Sometimes, as with the trash-can moms of 1997, a media circus removes any possibility of leniency, but even these young women were paroled after three years in jail.

It may seem odd to call the protection of identifiable human life a "taboo," because it seems self-evident. The very act of holding the sacredness of life up to the light to examine it appears to be monstrous. But that reaction is precisely what makes a taboo a taboo, and questioning the identifiable-human-life taboo on intellectual and even moral grounds is certainly possible. In 1911 an English physician, Charles Mercier, presented arguments that infanticide should be considered a less heinous crime than the murder of an older child or an adult:

Like the nuclear taboo, the human life taboo is in general a very good thing. Consider this memoir from a man whose family was migrating with a group of settlers from California to Oregon in 1846. During their journey they came across an abandoned eight-year-old Native American girl, who was starving, naked, and covered with sores.

A council among the men was held to see what should be done with her. My father wanted to take her along; others wanted to kill her and put her out of her misery. Father said that would be willful murder. A vote was taken and it was decided to do nothing about it, but to leave her where we found her. My mother and my aunt were unwilling to leave the little girl. They stayed behind to do all they could for her. When they finally joined us their eyes were wet with tears. Mother said she had knelt down by the little girl and had asked God to take care of her. One of the young men in charge of the horses felt so badly about leaving her, he went back and put a bullet through her head and put her out of her misery.

The human life taboo was cemented by our reaction to the Nazi Holocaust, which proceeded in stages. It started with the euthanizing of mentally retarded people, psychiatric patients, and children with disabilities, then expanded to homosexuals, inconvenient Slavs, the Roma, and the Jews. Among the masterminds of the Holocaust and the citizens who were complicit with them, each stage may have made the next more thinkable. A bright line at the top of the slippery slope, we now reason, might have prevented people from sliding into depravity. Since the Holocaust a taboo on human manipulations of life and death have put public discussions of infanticide, eugenics, and active euthanasia beyond the pale. But like all taboos, the human life taboo is incompatible with certain features of reality, and fierce debates in bioethics today hinge on how to reconcile it with the fuzziness of the biological boundary that demarcates human life during embryogenesis, comas, and noninstantaneous deaths.

Today the story leaves us in shock. But in the moral universe of the settlers, allowing the girl to die and actively ending her life were live options. Though we engage in similar reasoning when we put an aging pet or a horse with a broken leg out of its misery, we place humans in a sacred category. Trumping all calculations based on empathy and mercy is a veto based on human life: an identifiable human's right to live is not negotiable.

For almost a millennium and a half the Judeo-Christian prohibition against infanticide coexisted with massive infanticide in practice. According to one historian, exposure of infants during the Middle Ages "was practiced on a gigantic scale with absolute impunity, noticed by writers with most frigid indifference." Milner cites birth records showing an average of 5.1 births among wealthy families, 2.9 among the middle class, and 1.8 among the poor, adding, "There was no evidence that the number of pregnancies followed similar lines." In 1527 a French priest wrote that "the latrines resound with the cries of children who have been plunged into them."

Any taboo that contravenes powerful inclinations in human nature must be fortified with layers of euphemism and hypocrisy, and it may have little practical effect on the proscribed activity. That is what happened with infanticide in most of European history. Perhaps the least contentious claim about human nature is that humans are apt to have sex under a wider range of circumstances than those in which they are capable of bringing up the resulting babies. In the absence of contraception, abortion, or an elaborate system of social welfare, many children will be born without suitable caregivers to bring them to adulthood. Taboo or no taboo, many of those newborns will end up dead.

At various points in the late Middle Ages and the early modern period, systems of criminal justice tried to do something about infanticide. The steps they took were a dubious improvement. In some countries, the breasts of unmarried servant women were regularly inspected for signs of milk, and if the woman could not produce a baby, she would be tortured to find out what happened to it. A woman who concealed the birth of a baby who did not survive was presumed guilty of infanticide and put to death, often by being sewn into a sack with a couple of feral cats and thrown into a river. Even with less colorful methods of punishment, the campaign to reduce infanticide by executing young mothers, many of them servants impregnated by the man of the house, began to tug on people's consciences, as they realized they were preserving the sanctity of human life by allowing men to dispose of their inconvenient mistresses.

Various fig leaves were procured. The phenomenon of "overlying," in which a mother would accidentally smother an infant by rolling over it in her sleep, at times became an epidemic. Women were invited to drop off their unwanted babies at foundling homes, some of them equipped with turntables and trapdoors to ensure anonymity. The mortality rates for the inhabitants of these h omes ranged from 50 percent to more than 99 percent. 149 Women handed over their infants to wet nurses or "baby farmers" who were known to have similar rates of success. Elixirs of opium, alcohol, and treacle were readily obtainable by mothers and wet nurses to becalm a cranky infant, and at the right dosage it could becalm them very effectively indeed. Many a child who survived infancy was sent to a workhouse, "without the inconvenience of too much food or too much clothing," as Dickens described them in Oliver Twist, and where "it did perversely happen in eight and a half cases out of ten, either that it sickened from want and cold, or fell into the fire from neglect, or got half-smothered by accident; in any one of which cases, the miserable little being was usually summoned into another world, and there gathered to the fathers it had never known in this." Even with these contrivances, tiny corpses were a frequent sight in parks, under bridges, and in ditches. According to a British coroner in 1862, "The police seemed to think no more of finding a dead child than they did of finding a dead cat or a dead dog."

In the 20th century, even before abortions were widely available, a girl who got pregnant was less likely to give birth alone and secretly kill her newborn, because other people had set up alternatives, such as homes for unwed mothers, orphanages that were not death camps, and agencies that found adoptive and foster parents for motherless children. Why did governments, charities, and religions start putting money into these lifesavers? One gets a sense that children became more highly valued, and that our collective circle of concern has widened to embrace their interests, beginning with their interest in staying alive. A look at other aspects of the treatment of children confirms that the recent changes have been sweeping.

The several-thousandfold reduction in infanticide enjoyed in the Western world today is partly a gift of affluence, which leaves fewer mothers in desperate straits, and partly a gift of technology, in the form of safe and reliable contraception and abortion that has reduced the number of unwanted newborns. But it also reflects a change in the valuation of children. Rather than leaving it a pious aspiration, societies finally made good on the doctrine that the lives of infants are sacred -- regardless of who bore them, regardless of how shapeless and foul they were at birth, regardless of how noticeable a gap their loss would leave in a family circle, regardless of how expensive they were to feed and care for.

It is true that in much of the world today, a similar proportion of pregnancies end in abortion as the fraction that in centuries past ended in infanticide. 151 Women in the developed West abort between 12 and 25 percent of their pregnancies; in some of the former communist countries the proportion is greater than half. In 2003 a million fetuses were aborted in the United States, and about 5 million were aborted throughout Europe and the West, with at least another 11 million aborted elsewhere in the world. If abortion counts as a form of violence, the West has made no progress in its treatment of children. Indeed, because effective abortion has become widely available only since the 1970s (especially, in the United States, with the 1973 Roe v. Wade Supreme Court decision), the moral state of the West hasn't improved; it has collapsed.

Before turning to the bigger picture of the appreciation of children in the West, I must spend a few words on a more jaundiced view of the historical fate of infanticide. According to an alternative history, the major long-term trend in the West is that people have switched from killing children shortly after they are born to killing them shortly after they are conceived.

This is not the place to discuss the morality of abortion, but the larger context of trends in violence can provide some insight into how people conceive of abortion. Many opponents of legalized abortion predicted that acceptance of the practice would cheapen human life and put society on a slippery slope toward infanticide, euthanasia of the handicapped, a devaluation of the lives of children, and eventually widespread murder and genocide. Today we can say with confidence that that has not happened. Though abortion has been available in most of the Northern Hemisphere for decades, no country has allowed the deadline for abortions during pregnancy to creep steadily forward into legal infanticide, nor has the availability of abortion prepared the ground for euthanasia of disabled children. Between the time when abortion was made widely available and today, the rate of every category of violence has gone down, and, as we shall see, the valuation of the lives of children has shot up.

Opponents of abortion may see the decline in every form of violence but the killing of fetuses as a stunning case of moral hypocrisy. But there is another explanation for the discrepancy. Modern sensibilities have increasingly conceived moral worth in terms of consciousness, particularly the ability to suffer and flourish, and have identified consciousness with the activity of the brain. The change is a part of the turning away from religion and custom and toward science and secular philosophy as a source of moral illumination. Just as the legally recognized end of life is now defined by the cessation of brain activity rather than the cessation of a heartbeat, the beginning of life is sensed to depend on the first stirrings of consciousness in the fetus. The current understanding of the neural basis of consciousness ties it to reverberating neural activity between the thalamus and the cerebral cortex, which begins at around twenty-six weeks of gestational age. More to the point, people conceive of fetuses as less than fully conscious: the psychologists Heather Gray, Kurt Gray, and Daniel Wegner have shown that people think of fetuses as more capable of experience than robots or corpses, but less capable than animals, babies, children, and adults. The vast majority of abortions are carried out well before the milestone of having a functioning brain, and thus are safely conceptualized, according to this understanding of the worth of human life, as fundamentally different from infanticide and other forms of violence.

At the same time, we might expect a general distaste for the destruction of any kind of living thing to turn people away from abortion even when they don't equate it with murder. And that indeed has happened. It's a little-known fact that rates of abortion are falling throughout the world. Figure 7-16 shows the rates of abortion in the major regions in which data are available (albeit differing widely in quality) in the 1980s, 1996, and 2003.

The decline has been steepest in the countries of the former Soviet bloc, which were said to have had a "culture of abortion." During the communist era abortions were readily available, but contraceptives, like every other consumer good, were allocated by a central commissar rather than by supply and demand, so they were always in short supply. But abortions have also become less common in China, the United States, and the Asian and Islamic countries in which they are legal. Only in India and Western Europe did abortion rates fail to decline, and those are the regions where the rates were lowest to begin with.

FIGURE 7-16: Abortions in the world, 1980-2003

Sources: 1980s: Henshaw, 1990; 1996 & 2003: Sedgh et al., 2007. "Eastern Europe" comprises Bulgaria, Czechoslovakia/Czech Republic & Slovakia, Hungary, Yugoslavia/Serbia-Montenegro, Romania. "Western Europe" comprises Belgium, Denmark, England and Wales, Finland, the Netherlands, Norway, Scotland, Sweden. "Asia" comprises Singapore, Japan, South Korea (2003 equated with 1996). "Islamic" comprises Tunisia and Turkey.

The causes of much of the decline, to be sure, are practical. Contraception is cheaper and more convenient than abortion, and if it's readily available it will be the first choice of people with the foresight and self-control to use it. But presumably abortion has a moral dimension even among those who undergo it and among their compatriots who want to keep the option safe and legal. Abortion is seen as something to be minimized, even if it is not criminalized. If so, the trends in abortion offer a sliver of common ground in the rancorous debate between the so-called pro-life and pro-choice factions. The countries that allow abortion have not let an indifference to life put them on a slippery slope to infanticide or other forms of violence. But these same countries increasingly act as if abortion is undesirable, and they may be reducing its incidence as part of the move to protect all living things.

During the long, sad history of violence against children, even when infants survived the day of their birth it was only to endure harsh treatment and cruel punishments in the years to come. Though hunter-gatherers tend to use corporal punishment in moderation, the dominant method of child-rearing in every other society comes right out of Alice in Wonderland: "Speak roughly to your little boy, and beat him when he sneezes." The reigning theory of child development was that children were innately depraved and could be socialized only by force. The expression "Spare the rod and spoil the child" has been attributed to an advisor to the king of Assyria in the 7th century BCE and may have been the source of Proverbs 13:24, "He that spareth the rod hateth his son: But he that loveth him chasteneth him betimes." A medieval French verse advised, "Better to beat your child when small than to see him hanged when grown." The Puritan minister Cotton Mather (Increase's son) extended the concern for the child's well-being to the hereafter: "Better whipt than Damn'd."

Severe corporal punishment was common for centuries. One survey found that in the second half of the 18th century, 100 percent of American children were beaten with a stick, whip, or other weapon. Children were also liable to punishment by the legal system; a recent biography of Samuel Johnson remarks in passing that a seven-year-old girl in 18th-century England was hanged for stealing a petticoat. Even at the turn of the 20th century, German children "were regularly placed on a red-hot iron stove if obstinate, tied to their bedposts for days, thrown into cold water or snow to 'harden' them, [and] forced to kneel for hours every day against the wall on a log while the parents ate and read." During toilet training many children were tormented with enemas, and at school they were "beaten until [their] skin smoked."

That children with devils in them had to be beaten goes without saying. A panoply of beating instruments existed for that purpose, from cat-o'-nine tails and whips to shovels, canes, iron rods, bundles of sticks, the discipline (a whip made of small chains), the goad (shaped like a cobbler's knife, used to prick the child on the head or hands) and special school instruments like the flapper, which had a pear-shaped end and a round hole to raise blisters. The beatings described in the sources were almost always severe, involved bruising and bloodying of the body, began in infancy, were usually erotically tinged by being inflicted on bare parts of the body near the genitals and were a regular part of the child's daily life.

As with all punishments, human ingenuity rose to the technological challenge of delivering experiences that were as unpleasant as possible. DeMause writes of medieval Europe:

Children were subjected to psychological torture as well. Much of their entertainment was filled with reminders that they might be abandoned by parents, abused by stepparents, or mutilated by ogres and wild animals. Grimm's fairy tales were just a few of the advisories that may be found in children's literature of the misfortunes that can befall a careless or disobedient child. English babies, for example, were soothed to sleep with a lullaby about Napoleon:

The harsh treatment was not unique to Europe. The beating of children has been recorded in ancient Egypt, Sumeria, Babylonia, Persia, Greece, Rome, China, and Aztec Mexico, whose punishments included "sticking the child with thorns, having their hands tied and then being stuck with pointed agave leaves, whippings, and even being held over a fire of dried axi peppers and being made to inhale the acrid smoke." DeMause notes that well into the 20th century, Japanese children were subjected to "beating and burning of incense on the skin as routine punishments, cruel bowel training with constant enemas,… kicking, hanging by the feet, giving cold showers, strangling, driving a needle into the body, cutting off a finger joint." (A psychoanalyst as well as a historian, deMause had plenty of material with which to explain the atrocities of World War II.)

And he'll eat you, eat you, eat you,

Limb from limb at once he'll tear you,

And threw them out the window.

And he'll beat you, beat you, beat you,

Why would any parents torture, starve, neglect, and terrify their own children? One might naïvely think that parents would have evolved to nurture their children without stinting, since having viable offspring is the be-all and endall of natural selection. Children too ought to submit to their parents' guidance without resistance, since it is offered for their own good. The naïve view predicts a harmony between parent and child, since each "wants" the same thing -- for the child to grow up healthy and strong enough to have children of its own.

A recurring archetype in children's verse is the child who commits a minor slipup or is unjustly blamed for one, whereupon his stepmother butchers him and serves him for dinner to his unwitting father. In a Yiddish version, the victim of one such injustice sings posthumously to his sister:

Eaten by my father.

As he gallops past the house,

Every morsel, snap, snap, snap.

Murdered by my mother,

And Sheyndele, when they were done

And he'll beat you all to pap,

Sucked the marrow from my bones

Just as pussy tears a mouse.

Baby, baby, if he hears you,

It was Trivers who first noticed that the theory of natural selection predicts no such thing. Some degree of conflict between parent and offspring is rooted in the evolutionary genetics of the family. Parents have to apportion their investment (in resources, time, and risk) across all their children, born and unborn. All things being equal, every offspring is equally valuable, though each benefits from parental investment more when it is young and helpless than when it can fend for itself. The child sees things differently. Though an offspring has an interest in its siblings' welfare, since it shares half its genes with each full sib, it shares all of its genes with itself, so it has a disproportionate interest in its own welfare. The tension between what a parent wants (an equitable allocation of its worldly efforts to all its children) and what a child wants (a lopsided benefit to itself compared to its siblings) is called parent-offspring conflict. Though the stakes of the conflict are the parents' investment in a child and its siblings, those siblings need not yet exist: a parent must also conserve strength for future children and grandchildren. Indeed, the first dilemma of parenthood -- whether to keep a newborn -- is just a special case of parent-offspring conflict.

The theory of parent-offspring conflict says nothing about how much investment an offspring should want or how much a parent should be prepared to give. It says only that however much parents are willing to give, the offspring wants a bit more. Children cry when they are in need of help, and parents cannot ignore the cries. But children are expected to cry a bit louder and longer than their objective need calls for. Parents discipline children to keep them out of danger, and socialize them to be effective members of their community. But parents are expected to discipline children a bit more for their own convenience, and to socialize them to be a bit more accommodating to their siblings and kin, than the levels that would be in the interests of the children themselves. As always, the teleological terms in the explanation --"wants," "interests," "for"-- don't refer to literal desires in the minds of people, but are shorthand for the evolutionary pressures that shaped those minds.

Parent-offspring conflict explains why child-rearing is always a battle of wills. What it does not explain is why that battle should be fought with rods and birches in one era and lectures and time-outs in another. In retrospect, it's hard to avoid sorrow for the millennia of children who have needlessly suffered at the hands of their caregivers. Unlike the tragedy of war, where each side has to be as fierce as its adversary, the violence of child-rearing is entirely one-sided. The children who were whipped and burned in the past were no naughtier than the children of today, and they ended up no better behaved as adults. On the contrary, we have seen that the rate of impulsive violence of yesterday's adults was far higher than today's. What led the parents of our era to the discovery that they could socialize their children with a fraction of the brute force that was used by their ancestors?

The first nudge was ideological, and like so many other humanitarian reforms it originated in the Age of Reason and the Enlightenment. Children's tactics in parent-offspring conflict have led parents in every era to call them little devils. During the ascendancy of Christianity, that intuition was ratified by a religious belief in innate depravity and original sin. A German preacher in the 1520s, for instance, sermonized that children harbored wishes for "adultery, fornication, impure desires, lewdness, idol worship, belief in magic, hostility, quarreling, passion, anger, strife, dissension, factiousness, hatred, murder, drunkenness, gluttony," and he was just getting started. The expression "beat the devil out of him" was more than a figure of speech! Also, a fatalism about the unfolding of life made child development a matter of fate or divine will rather than the responsibility of parents and teachers.

One paradigm shift came from John Locke's Some Thoughts Concerning Education, which was published in 1693 and quickly went viral. Locke suggested that a child was "only as white Paper, or Wax, to be moulded and fashioned as one pleases'-- a doctrine also called the tabula rasa (scraped tablet) or blank slate. Locke wrote that the education of children could make "a great difference in mankind," and he encouraged teachers to be sympathetic toward their pupils and to try to take their viewpoints. Tutors should carefully observe the "change in temper" in their students and should help them enjoy their studies. And teachers should not expect young children to show the same "carriage, seriousness, or application" as older ones. On the contrary, "they must be permitted… the foolish and childish actions suitable to their years."

The idea that the way children are treated determines the kinds of adults they grow into is conventional wisdom today, but it was news at the time. Several of Locke's contemporaries and successors turned to metaphor to remind people about the formative years of life. John Milton wrote, "The childhood shows the man as morning shows the day." Alexander Pope elevated the correlation to causation: "Just as the twig is bent, the tree's inclined." And William Wordsworth inverted the metaphor of childhood itself: "The child is father of the man." The new understanding required people to rethink the moral and practical implications of the treatment of children. Beating a child was no longer an exorcism of malign forces possessing a child, or even a technique of behavior modification designed to reduce the frequency of bratty behavior in the present. It shaped the kind of person that the child would grow into, so its consequences, foreseen and unforeseen, would alter the makeup of civilization in the future.

Another gestalt shift came from Rousseau, who replaced the Christian notion of original sin with the romantic notion of original innocence. In his 1762 treatise Émile, or On Education, Rousseau wrote, "Everything is good as it leaves the hand of the Author of things, and everything degenerates in the hands of man." Foreshadowing the theories of the 20th-century psychologist Jean Piaget, Rousseau divided childhood into a succession of stages centered on Instinct, Sensations, and Ideas. He argued that young children have not yet reached the Age of Ideas, and so should not be expected to reason in the ways of adults. Rather than drilling youngsters in the rules of good and evil, adults should allow children to interact with nature and learn from their experiences. If in the course of exploring the world they damaged things, it was not from an intention to do harm but from their own innocence. "Respect childhood," he implored, and "leave nature to act for a long time before you get involved with acting in its place." The 19th-century Romantic movement inspired by Rousseau saw childhood as a period of wisdom, purity, and creativity, a stage that children should be left to enjoy rather than be disciplined out of. The sensibility is familiar today but was radical at the time.

During the Enlightenment, elite opinion began to incorporate the childfriendly doctrines of the blank slate and original innocence. But historians of childhood place the turning point in the actual treatment of children considerably later, in the decades surrounding the turn of the 20th century. The economist Viviana Zelizer has suggested that the era from the 1870s to the 1930s saw a "sacralization" of childhood among middle- and upper-class parents in the West. That was when children attained the status we now grant them: "economically worthless, emotionally priceless." The era was inaugurated in England when a "baby-farming" scandal led to the formation of the Infant Protection Society in 1870 and to the Infant Life Protection Acts of 1872 and 1897. Around the same time, pasteurization and sterilized bottles meant that fewer infants were outsourced to infanticidal wet nurses. Though the Industrial Revolution originally moved children from backbreaking labor on farms to backbreaking labor in mills and factories, legal reforms increasingly restricted child labor. At the same time, the affluence that flowed from the maturing Industrial Revolution drove rates of infant mortality downward, reduced the need for child labor, and provided a tax stream that could support social services. More children went to school, which soon became compulsory and free. To deal with the packs of urchins, ragamuffins, and artful dodgers who roamed city streets, child welfare agencies founded kindergartens, orphanages, reform schools, fresh-air camps, and boys' and girls' clubs. Stories for children were written to give them pleasure rather than to terrorize them or moralize to them. The Child Study movement aimed for a scientific approach to human development and began to replace the superstition and bunkum of old wives with the superstition and bunkum of child-rearing experts.

In Manhattan in 1874, the neighbors of ten-year-old Mary Ellen McCormack, an orphan being raised by an adoptive mother and her second husband, noticed suspicious cuts and bruises on the girl's body. They reported her to the Department of Public Charities and Correction, which administered the city's jails, poorhouses, orphanages, and insane asylums. Since there were no laws that specifically protected children, the caseworker contacted the American Society for the Protection of Animals. The society's founder saw an analogy between the plight of the girl and the plight of the horses he rescued from violent stable owners. He engaged a lawyer who presented a creative interpretation of habeas corpus to the New York State Supreme Court and petitioned to have her removed from her home. The girl calmly testified:

We have seen that during periods of humanitarian reform, a recognition of the rights of one group can lead to a recognition of others by analogy, as when the despotism of kings was analogized to the despotism of husbands, and when two centuries later the civil rights movement inspired the women's rights movement. The protection of abused children also benefited from an analogy -- in this case, believe it or not, with animals.

The New York Times reprinted the testimony in an article entitled "Inhumane Treatment of a Little Waif," and the girl was removed from the home and eventually adopted by her caseworker. Her lawyer set up the New York Society for the Prevention of Cruelty to Children, the first protective agency for children anywhere in the world. Together with other agencies founded in its wake, it set up shelters for battered children and lobbied for laws that punished their abusive parents. Similarly, in England the first legal case to protect a child against an abusive parent was taken up by the Royal Society for the Prevention of Cruelty to Animals, and out of it grew the National Society for the Prevention of Cruelty to Children.

Mamma has been in the habit of whipping and beating me almost every day. She used to whip me with a twisted whip -- a rawhide. I have now on my head two black-and-blue marks which were made by Mamma with the whip, and a cut on the left side of my forehead which was made by a pair of scissors in Mamma's hand… I never dared speak to anybody, because if I did I would get whipped.

But whatever the causes were, they did not stop in the 1930s. Benjamin Spock's perennial bestseller Baby and Child Care was considered radical in 1946 because it discouraged mothers from spanking their children, stinting on affection, and regimenting their routines. Though the indulgence of postwar parents was a novelty at the time (widely and spuriously blamed for the excesses of the baby boomers), it was by no means a high-water mark. When the boomers became parents, they were even more solicitous of their children. Locke, Rousseau, and the 19th-century reformers had set in motion an escalator of gentleness in the treatment of children, and in recent decades its rate of ascent has accelerated.

Though the rollover of the 19th century saw an acceleration in the valuation of children in the West, it was neither an abrupt transition nor a one-shot advance. Expressions of love of children, of grief at their loss, and of dismay at their mistreatment can be found in every period of European history and in every culture. Even many of the parents who treated their children cruelly were often laboring under superstitions that led them to think they were acting in the child's best interests. And as with many declines in violence, it's hard to disentangle all the changes that were happening at once -- enlightened ideas, increasing prosperity, reformed laws, changing norms.

Since 1950, people have become increasingly loath to allow children to become the victims of any kind of violence. The violence people can most easily control, of course, is the violence they inflict themselves, namely by spanking, smacking, slapping, paddling, birching, tanning, hiding, thrashing, and other forms of corporal punishment. Elite opinion on corporal punishment changed dramatically during the 20th century. Other than in fundamentalist Christian groups, it's rare today to hear people say that sparing the rod will spoil the child. Scenes of fathers with belts, mothers with hairbrushes, and teary children tying pillows to their bruised behinds are no longer common in family entertainment.

At least since Dr. Spock, child-care gurus have increasingly advised against spanking. Today every pediatric and psychological association opposes the practice, though not always in language as clear as the title of a recent article by Murray Straus: "Children Should Never, Ever, Be Spanked No Matter What the Circumstances." The expert opinion recommends against spanking for three reasons. One is that spanking has harmful side effects down the line, including aggression, delinquency, a deficit in empathy, and depression. The cause-and-effect theory, in which spanking teaches children that violence is a way to solve problems, is debatable. Equally likely explanations for the correlation between spanking and violence are that innately violent parents have innately violent children, and that cultures and neighborhoods that tolerate spanking also tolerate other kinds of violence. The second reason not to spank a child is that spanking is not particularly effective in reducing misbehavior compared to explaining the infraction to the child and using nonviolent measures like scolding and time-outs. Pain and humiliation distract children from pondering what they did wrong, and if the only reason they have to behave is to avoid these penalties, then as soon as Mom's and Dad's backs are turned they can be as naughty as they like. But perhaps the most compelling reason to avoid spanking is symbolic. Here is Straus's third reason why children should never, ever be spanked: "Spanking contradicts the ideal of nonviolence in the family and society."

Have parents been listening to the experts, or perhaps coming to similar conclusions on their own? Public opinion polls sometimes ask people whether they agree with statements like "It is sometimes necessary to discipline a child with a good, hard spanking" or "There are certain circumstances when it is all right to smack a child." The level of agreement depends on the wording of the question, but in every poll in which the same question has been asked in different years, the trend is downward. Figure 7-17 shows the trends since 1954 from three American datasets, together with surveys from Sweden and New Zealand. Before the early 1980s, around 90 percent of respondents in the English-speaking countries approved of spanking. In less than a generation, the percentage had fallen in some polls to just more than half. The levels of approval depend on the country and region: Swedes approve of spanking far less than do Americans or Kiwis, and Americans themselves are diverse, as we would expect from the southern culture of honor. In a 2005 survey, spanking approval rates ranged from around 55 percent in northern blue states (those that tend to vote for Democrats), like Massachusetts and Vermont, to more than 85 percent in southern red states (those that tend to vote for Republicans), like Alabama and Arkansas. Across the fifty states, the rate of approval of spanking tracks the homicide rate (the two measures show a correlation of 0.52 on a scale from -1 to 1), which could mean that spanked children grow up to be killers, but more likely that subcultures that encourage the spanking of children also encourage the violent defense of honor among adults. But every region showed a decline, so that by 2006 the southern states disapproved of spanking in the same proportion that the north-central and mid-Atlantic states did in 1986.

FIGURE 7-17: Approval of spanking in the United States, Sweden, and New Zealand, 1954-2008

Sources: Gallup/ABC: Gallup, 1999; ABC News, 2002. Straus: Straus, 2001, p. 206. General Social Survey: http://www. norc. org/GSS+Website/, weighted means. New Zealand: Carswell, 2001. Sweden: Straus, 2009.

What about actual behavior? Many parents will still slap a toddler's hand if the child reaches for a forbidden object, but in the second half of the 20th century every other kind of corporal punishment declined. In the 1930s American parents spanked their children more than 3 times a month, or more than 30 times a year. By 1975 the figure had fallen to 10 times a year and by 1985 to around 7. Even steeper declines were seen in Europe. In the 1950s, 94 percent of Swedes spanked their children, and 33 percent did so every day; by 1995, the figures had plunged to 33 and 4 percent. By 1992, German parents had come a long way from their great-grandparents who had placed their grandparents on hot stoves and tied them to bedposts. But 81 percent still slapped their children on the face, 41 percent spanked them with a stick, and 31 percent beat them to the point of bruising. By 2002, these figures had sunk to 14 percent, 5 percent, and 3 percent.

In 1979 the government of Sweden outlawed spanking altogether. The other Scandinavian countries soon joined it, followed by several countries of Western Europe. The United Nations and the European Union have called on all their member nations to abolish spanking. Several countries have launched public awareness campaigns against the practice, and twenty-four have now made it illegal.

There remains a lot of variation among countries today. No more than 5 percent of college students in Israel, Hungary, the Netherlands, Belgium, and Sweden recall being hit as a teenager, but more than a quarter of the students in Tanzania and South Africa do. In general, wealthier countries spank their children less, with the exception of developed Asian nations like Taiwan, Singapore, and Hong Kong. The international contrast is replicated among ethnic groups in the United States, where African Americans and Asians spank more than whites. But the level of approval of spanking has declined in all three groups.

The prohibition of spanking represents a stunning change from millennia in which parents were considered to own their children, and the way they treated them was considered no one else's business. But it is consistent with other intrusions of the state into the family, such as compulsory schooling, mandatory vaccination, the removal of children from abusive homes, the imposition of lifesaving medical care over the objections of religious parents, and the prohibition of female genital cutting by communities of Muslim immigrants in European countries. In one frame of mind, this meddling is a totalitarian imposition of state power into the intimate sphere of the family. But in another, it is part of the historical current toward a recognition of the autonomy of individuals. Children are people, and like adults they have a right to life and limb (and genitalia) that is secured by the social contract that empowers the state. The fact that other individuals -- their parents -- stake a claim of ownership over them cannot negate that right.

American sentiments tend to weight family over government, and currently no American state prohibits corporal punishment of children by their parents. But when it comes to corporal punishment of children by the government, namely in schools, the United States has been turning away from this form ofof violence. Even in red states, where three-quarters of the people approve of spanking by parents, only 30 percent approve of paddling in schools, and in the blue states the approval rate is less than half that. And since the 1950s the level of approval of corporal punishment in schools has been in decline (figure 7-18). The growing disapproval has been translated into legislation. Figure 7-19 shows the shrinking proportion of American states that still allow corporal punishment in schools.

Sources: Data for 1954-94 from Gallup, 1999; data for 2002 from ABC News, 2002.

FIGURE 7-18: Approval of corporal punishment in schools in the United States, 1954-2002

Source: Data from Leiter, 2007.

The trend is even more marked in the international arena, where corporal punishment in schools is now seen as a violation of human rights, like other forms of extrajudicial government violence. It has been condemned by the UN Committee on the Rights of the Child, the UN Human Rights Committee, and the UN Committee Against Torture, and has been banned by 106 countries, more than half of the world's total.

Though a majority of Americans still endorse corporal punishment by parents, they draw an increasingly sharp line between mild violence they consider discipline, such as spanking and slapping, and severe violence they consider abuse, such as punching, kicking, whipping, beating, and terrorizing (for example, threatening a child with a knife or gun, or dangling it over a ledge). In his surveys of domestic violence, Straus gave respondents a checklist that included punishments that are now considered abusive. He found that the number of parents admitting to them almost halved between 1975 and 1992, from 20 percent of mothers to a bit more than 10 percent.

FIGURE 7-19: American states allowing corporal punishment in schools, 1954-2010

A problem in self-reports of violence by perpetrators (as opposed to self-reports by victims) is that a positive response is a confession to a wrongdoing. An apparent decline in parents beating their children may really be a decline in parents owning up to it. At one time a mother who left bruises on her child might consider it within the range of acceptable discipline. But starting in the 1980s, a growing number of opinion leaders, celebrities, and writers of television dramas began to call attention to child abuse, often by portraying abusive parents as reprehensible ogres or grown-up children as permanently scarred. In the wake of this current, a parent who bruised a child in anger might keep her mouth shut when the surveyor called. We do know that child abuse had become more of a stigma in this interval. In 1976, when people were asked, "Is child abuse a serious problem in this country?" 10 percent said yes; when the same question was asked in 1985 and 1999, 90 percent said yes. Straus argued that the downward trend in his violence survey captured both a decline in the acceptance of abuse and a decline in actual abuse; even if much of the decline was in acceptance, he added, that would be something to celebrate. A decreasing tolerance of child abuse led to an expansion in the number of abuse hotlines and child protection officers, and to an expanded mandate among police, social workers, school counselors, and volunteers to look for signs of abuse and take steps that would lead to abusers being punished or counseled and to children being removed from the worst homes.

FIGURE 7-20: Child abuse in the United States, 1990-2007

Have the changes in norms and institutions done any good? The National Child Abuse and Neglect Data System was set up to aggregate data on substantiated cases of child abuse from child protection agencies around the country. The psychologist Lisa Jones and the sociologist David Finkelhor have plotted their data over time and shown that from 1990 to 2007 the rate of physical abuse of children fell by half (figure 7-20).

Sources: Data from Jones & Finkelhor, 2007; see also Finkelhor & Jones, 2006.

Jones and Finkelhor also showed that during this time the rate of sexual abuse, and the incidence of violent crimes against children such as assault, robbery, and rape, also fell by a third to two-thirds. They corroborated the declining numbers with sanity checks such as victimization surveys, homicide data, offender confessions, and rates of sexually transmitted diseases, all of which are in decline. In fact over the past two decades the lives of children and adolescents improved in just about every way you can measure. They were also less likely to run away, to get pregnant, to get into trouble with the law, and to kill themselves. England and Wales have also enjoyed a decline in violence against children: a recent report has shown that since the 1970s, the rate of violent deaths of children fell by almost 40 percent.

The decline of child abuse in the 1990s coincided in part with the decline of adult homicide, and its causes are just as hard to pinpoint. Finkelhor and Jones examined the usual suspects. Demography, capital punishment, crack cocaine, guns, abortion, and incarceration can't explain the decline. The prosperity of the 1990s can explain it a little, but can't account for the decline in sexual abuse, nor a second decline of physical abuse in the 2000s, when the economy was in the tank. The hiring of more police and interveners from social service agencies probably helped, and Finkelhor and Jones speculate that another exogenous factor may have made a difference. The early 1990s was the era of Prozac Nation and Running on Ritalin. The massive expansion in the prescription of medication for depression and attention deficit disorder may have lifted many parents out of depression and helped many children control their impulses. Finkelhor and Jones also pointed to nebulous but potentially potent changes in cultural norms. The 1990s, as we saw in chapter 3, hosted a civilizing offensive that reversed some of the licentiousness of the 1960s and made all forms of violence increasingly repugnant. And the Oprahfication of America applied a major stigma to domestic violence, while destigmatizing -- indeed beatifying -- the victims who brought it to light.

Until recently, adults had written off bullying as one of the trials of childhood. "Boys will be boys," they said, figuring that an ability to deal with intimidation in childhood was essential training for the ability to deal with it in adulthood. The victims, for their part, had nowhere to turn, because complaining to a teacher or parent would brand them as snitches and pantywaists and make their lives more hellish than ever.

Another kind of violence that torments many children is the violence perpetrated against them by other children. Bullying has probably been around for as long as children have been around, because children, like many juvenile primates, strive for dominance in their social circle by demonstrating their mettle and strength. Many childhood memoirs include tales of cruelty at the hands of other children, and the knuckle-dragging bully is a staple of popular culture. The rogues' gallery includes Butch and Woim in Our Gang, Biff Tannen in the Back to the Future trilogy, Nelson Muntz in The Simpsons, and Moe in Calvin and Hobbes (figure 7-21).

But in another of those historical gestalt shifts in which a category of violence flips from inevitable to intolerable, bullying has been targeted for elimination. The movement emerged from the ball of confusion surrounding the 1999 Columbine High School massacre, as the media amplified one another's rumors about the causes -- Goth culture, jocks, antidepressants, video games, Internet use, violent movies, the rock singer Marilyn Manson -- and one of them was bullying. As it turned out, the two assassins were not, as the media endlessly repeated, Goths who had been picked on by jocks. But a popular understanding took hold that the massacre was an act of revenge, and childhood professionals parlayed the urban legend into a campaign against bullying. Fortunately, the theory -- bully victim today, cafeteria sniper tomorrow -- coexisted with more respectable rationales, such as that victims of bullies suffer from depression, impaired performance in school, and an elevated risk of suicide. Currently forty-four states have laws that prohibit bullying in school, and many have mandatory curricula that denounce bullying, encourage empathy, and instruct children in how to resolve their conflicts constructively. Organizations of pediatricians and child psychologists have issued statements calling for prevention efforts, and magazines, television programs, the Oprah Winfrey empire, and even the president of the United States have targeted it as well. In another decade, the facetious treatment of bullying in the Calvin and Hobbes cartoon may become as offensive as the spank-the-wife coffee ads from the 1950s are to us today.

Psychological consequences aside, the moral case against bullying is ironclad. As Calvin observed, once you grow up, you can't go beating people up for no reason. We adults protect ourselves with laws, police, workplace regulations, and social norms, and there is no conceivable reason why children should be left more vulnerable, other than laziness or callousness in considering what life is like from their point of view. The increased valuation of children, and the universalizing of moral viewpoints of which it is a part, made the campaign to protect children from violence by their peers inevitable. So too the effort to protect them from other depredations. Children and teenagers have long been victims of petty crimes like the theft of lunch money, vandalism of their possessions, and sexual groping, which fall in the cracks between school regulations and criminal law enforcement. Here too the interests of younger humans are increasingly being recognized.

FIGURE 7-21: Another form of violence against children

Has it made a difference? It has begun to do so. In 2004 the U. S. Departments of Justice and Education issued a report on Indicators of School Crime and Safety that used victimization surveys and school and police statistics to document trends in violence against students from 1992 to 2003. The survey asked about bullying only in the last three years, but they tracked other kinds of violence for the entire period, and found that fighting, fear at school, and crimes such as theft, sexual assault, robbery, and assault all ramped downward, as figure 7-22 shows.

And contrary to yet another scare that has recently been ginned up by the media, based on widely circulated YouTube videos of female teenagers pummeling one another, the nation's girls have not gone wild. The rates of murder and robbery by girls are at their lowest level in forty years, and rates of weapon possession, fights, assaults, and violent injuries by and toward girls have been declining for a decade. With the popularity of YouTube, we can expect more of these video-driven moral panics (sadist grannies? bloodthirsty toddlers? killer gerbils?) in the years to come.

It is premature to say that the kids are all right, but they are certainly far better off than they used to be. Indeed, in some ways the effort to protect children against violence has begun to overshoot its target and is veering into the realm of sacrament and taboo.

FIGURE 7-22: Violence against youths in the United States, 1992-2003

Source: Data from DeVoe et al., 2004.

One of these taboos is what the psychologist Judith Harris calls the Nurture Assumption. Locke and Rousseau helped set off a revolution in the conceptualization of child-rearing by rewriting the role of caregivers from beating bad behavior out of children to shaping the kind of people they would grow into. By the late 20th century, the idea that parents can harm their children by abusing and neglecting them (which is true) grew into the idea that parents can mold their children's intelligence, personalities, social skills, and mental disorders (which is not). Why not? Consider the fact that children of immigrants end up with the accent, values, and norms of their peers, not of their parents. That tells us that children are socialized in their peer group rather than in their families: it takes a village to raise a child. And studies of adopted children have found that they end up with personalities and IQ scores that are correlated with those of their biological siblings but uncorrelated with those of their adopted siblings. That tells us that adult personality and intelligence are shaped by genes, and also by chance (since the correlations are far from perfect, even among identical twins), but are not shaped by parents, at least not by anything they do with all their children. Despite these refutations, the Nurture Assumption developed a stranglehold on professional opinion, and mothers have been advised to turn themselves into round-the-clock parenting machines, charged with stimulating, socializing, and developing the characters of the little blank slates in their care.

Another sacrament is the campaign to quarantine children from the slightest shred of a trace of a hint of a reminder of violence. In Chicago in 2009, after twenty-five students aged eleven to fifteen took part in the age-old sport of a cafeteria food fight, they were rounded up by the police, handcuffed, herded into a paddy wagon, photographed for mug shots, and charged with reckless conduct. Zero-tolerance policies for weapons on school property led to a threat of reform school for a six-year-old Cub Scout who had packed an all-in-one camping utensil in his lunch box, the expulsion of a twelve-year-old girl who had used a utility knife to cut windows out of a paper house for a class project, and the suspension of an Eagle Scout who followed the motto "Be Prepared" by keeping a sleeping bag, drinking water, emergency food, and a two-inch pocketknife in his car. Many schools have hired whistle-wielding "recess coaches" to steer children into constructive organized games, because left on their own they might run into one another, bicker over balls and jump ropes, or monopolize patches of playground.

condition, or gender.

Costumes may not demean any race, religion, nationality, handicapped

They should not depict gangs or horror characters, or be scary.

Adults have increasingly tried to keep depictions of violence out of children's culture. In a climactic sequence of the 1982 movie E. T., Elliott sneaks past a police roadblock with E. T. in the basket of his bicycle. When the film was rereleased in a 20th-anniversary version in 2002, Steven Spielberg had digitally disarmed the officers, using computer-generated imagery to replace their rifles with walkie-talkies. Around Halloween, parents are now instructed to dress their children in "positive costumes" such as historical figures or items of food like carrots or pumpkins rather than zombies, vampires, or characters from slasher films. A memo from a Los Angeles school carried the following costume advisory:

No fake fingernails.

Masks are allowed only during the parade.

No weapons, even fake ones.

Elsewhere in California, a mother who thought that her children might be frightened by the Halloween tombstones and monsters in a neighbor's yard called the police to report it as a hate crime.

The historical increase in the valuation of children has entered its decadent phase. Now that children are safe from being smothered on the day they are born, starved in foundling homes, poisoned by wet nurses, beaten to death by fathers, cooked in pies by stepmothers, worked to death in mines and mills, felled by infectious diseases, and beaten up by bullies, experts have racked their brains for ways to eke infinitesimal increments of safety from a curve of diminishing or even reversing returns. Children are not allowed to be outside in the middle of the day (skin cancer), to play in the grass (deer ticks), to buy lemonade from a stand (bacteria on lemon peel), or to lick cake batter off spoons (salmonella from uncooked eggs). Lawyer-vetted playgrounds have had their turf padded with rubber, their slides and monkey bars lowered to waist height, and their seesaws removed altogether (so that the kid at the bottom can't jump off and watch the kid at the top come hurtling to the ground -- the most fun part of playing on a seesaw). When the producers of Sesame Street issued a set of DVDs containing classic programs from the first years of the series (1969-74), they included a warning on the box that the shows were not suitable for children! The programs showed kids engaging in dangerous activities like climbing on monkey bars, riding tricycles without helmets, wriggling through pipes, and accepting milk and cookies from kindly strangers. Censored altogether was Monsterpiece Theater, because at the end of each episode the ascotted, smoking-jacketed host, Alistair Cookie (played by Cookie Monster), gobbled down his pipe, which glamorizes the use of tobacco products and depicts a choking hazard.

But nothing has transformed childhood as much as the risk of kidnapping by strangers, a textbook case in the psychology of fear. Since 1979, when six-year-old Etan Patz disappeared on his way to a school bus stop in lower Manhattan, kidnapped children have riveted the nation's attention, thanks to three interest groups that are dedicated to sowing panic among the nation's parents. The grief-stricken parents of murdered children understandably want something good to come out of their tragedies, and several have devoted their lives to raising awareness of child abductions. (One of them, John Walsh, campaigned to have photographs of missing children featured on milk cartons, and hosted a lurid television program, America's Most Wanted, which specialized in horrific kidnap-murders.) Politicians, police chiefs, and corporate publicists can smell a no-lose campaign from a mile away -- who could be against protecting children from perverts?-- and have held ostentatious ceremonies to announce protective measures named after missing children (Code Adam, Amber Alerts, Megan's Law, the National Missing Children Day). The media too can recognize a ratings pump when they see one, and have stoked the fear with round-the-clock vigils, documentaries in constant rotation ("It is every parent's nightmare…"), and a Law and Order spinoff dedicated to nothing but sex crimes.

Childhood has never been the same. American parents will not let their children out of their sight. Children are chauffeured, chaperoned, and tethered with cell phones, which, far from reducing parents' anxiety, only sends them into a tizzy if a child doesn't answer on the first ring. Making friends in the playground has given way to mother-arranged playdates, a phrase that didn't exist before the 1980s. Forty years ago two-thirds of children walked or biked to school; today 10 percent do. A generation ago 70 percent of children played outside; today the rate is down to 30 percent. In 2008 the nine-year-old son of the journalist Lenore Skenazy begged her to let him go home by himself on the New York subway. She agreed, and he made it home without incident. When she wrote about the vignette in a New York Sun column, she found herself at the center of a media frenzy in which she was dubbed "America's Worst Mom." (Sample headline: "Mom Lets 9-Year-Old Take Subway Home Alone: Columnist Stirs Controversy with Experiment in Childhood Independence.") In response she started a movement -- Free-Range Children -- and proposed National Take Our Children to the Park and Leave Them There Day, intended to get children to learn to play by themselves without constant adult supervision.

Skenazy is not, in fact, America's worst mom. She simply did what no politician, policeman, parent, or producer ever did: she looked up the facts. The overwhelming majority of milk-carton children were not lured into vans by sex perverts, child traffickers, or ransom artists, but were teenagers who ran away from home, or children taken by a divorced parent who was embittered by an unfavorable custody ruling. The annual number of abductions by strangers has ranged from 200 to 300 in the 1990s to about 100 today, around half of whom are murdered. With 50 million children in the United States, that works out to an annual homicide rate of one in a million (0.001 per hundred thousand, to use our usual metric). That's about a twentieth of the risk of drowning and a fortieth of the risk of a fatal car accident. The writer Warwick Cairns calculated that if you wanted your child to be kidnapped and held overnight by a stranger, you'd have to leave the child outside and unattended for 750,000 years.

One might reply that the safety of a child is so precious that even if these precautions saved a few lives a year, they would be worth the anxiety and expense. But the reasoning is spurious. People inescapably trade off safety for other good things in life, as when they set aside money for their children's college education rather than installing a sprinkler system in their homes, or drive with their children to a vacation destination rather than letting them play video games in the safety of their bedrooms all summer. The campaign for perfect safety from abductions ignores costs like constricting childhood experience, increasing childhood obesity, instilling chronic anxiety in working women, and scaring young adults away from having children.

And even if minimizing risk were the only good in life, the innumerate safety advisories would not accomplish it. Many measures, like the milk-carton wanted posters, are examples of what criminologists call crime-control theater: they advertise that something is being done without actually doing anything. When 300 million people change their lives to reduce a risk to 50 people, they will probably do more harm than good, because of the unforeseen consequences of their adjustments on the vastly more than 50 people who are affected by them. To take just two examples, more than twice as many children are hit by cars driven by parents taking their children to school as by other kinds of traffic, so when more parents drive their children to school to prevent them from getting killed by kidnappers, more children get killed. And one form of crime-control theater, electronic highway signs that display the names of missing children to drivers on freeways, may cause slowdowns, distracted drivers, and the inevitable accidents.

The movement over the past two centuries to increase the valuation of children's lives is one of the great moral advances in history. But the movement over the past two decades to increase the valuation to infinity can lead only to absurdities.